Computer software systems perform a variety of functions and include many diverse user interfaces that enable users to interact with the system. Different users can speak multiple languages, so that software systems often include distinct, non-English language versions of the software, each version having user interfaces that can interact with the user, that is, display instructions, information, data and other items, in one of several languages, such as French, Chinese, German, etc. Each particular language version of the software can be known as a “localized product”, that is, a version of the software interacting with a user in the Chinese language is the Chinese “localized product”.
Currently developers of these “multi-lingual” software systems with more than one localized product need to test each linguistic version separately and thoroughly, even though the underlying functions and processes of the software system are the same for the different languages. For example, in automated Graphical User Interface (GUI) testing of localized products, testing is done multiple times, one or more times for each of the several languages. Currently, each test in a specific language is performed with a test script in localized words. To obtain those localized test scripts, a scriptwriter must repeat the script creation job for each of the several languages by running each localized application. However, one problem with current technologies is that preparing multiple localized test scripts is time consuming and error-prone.
It would be desirable to provide a method for testing multi-lingual software in each of the multiple languages that can be performed without a separate test script for each language.